Heat Transfer Estimation in Flow Boiling of R134a within Microfin Tubes - Development of Explainable Machine Learning-Based Pipelines

Aug 4, 2026

Upload by Aria

The present study identifies suitable sequences of machine-learning models and promising input variables for estimating heat transfer during evaporating R134a flow in microfin tubes. Using available experimental data, dimensionless features characterizing evaporation are generated and provided to a machine-learning model, followed by feature selection and algorithm optimization. The selection approach reduces the input to six dimensionless parameters and improves interpretability by mapping them to governing physical phenomena. The optimized algorithm pipeline achieves MARD of 8.84% on the test set, outperforming the most accurate empirical model with MARD of 19.7%.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages24

Reading Time1 h 0 min

AccessFree to download

Introduction

Industrial relevance and motivation

Challenges for analytical and numerical methods

Role of microfin tubes and experimental correlations

Prior studies on evaporation and correlations

energies Article Heat Transfer Estimation in Flow Boiling of R134a within Microfin Tubes: Development of Explainable Machine Learning-Based Pipelines Shayan Milani, Keivan Ardam, Farzad Dadras Javan, Behzad Najafi *, Andrea Lucchini, Igor Matteo Carraretto and Luigi Pietro Maria Colombo Citation: Milani, S.; Ardam, K.; Dadras Javan, F.; Najafi, B.; Lucchini, A.; Carraretto, I.M.; Colombo, L.P.M. Heat Transfer Estimation in Flow Boiling of R134a within Microfin Tubes: Development of Explainable Machine Learning-Based Pipelines. Energies 2024, 17, 4074. [https://](https://)[ ](https://)[doi.org/10.3390/en17164074](doi.org/10.3390/en17164074) Academic Editor: Magdalena Piasecka Received: 13 June 2024 Revised: 2 August 2024 Accepted: 13 August 2024 Published: 16 August 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . Dipartimento di Energia, Politecnico di Milano, Via Lambruschini 4, 20156 Milano, Italy; [shayan.milani@mail.polimi.it](shayan.milani@mail.polimi.it) (S.M.); [keivan.ardam@mail.polimi.it](keivan.ardam@mail.polimi.it) (K.A.); [farzad.dadras@polimi.it](farzad.dadras@polimi.it) (F.D.J.); [andrea.lucchini@polimi.it](andrea.lucchini@polimi.it) (A.L.); [igormatteo.carraretto@polimi.it](igormatteo.carraretto@polimi.it) (I.M.C.); [luigi.colombo@polimi.it](luigi.colombo@polimi.it) (L.P.M.C.) * [Correspondence: behzad.najafi@polimi.it](Correspondence: behzad.najafi@polimi.it); Tel.: +39-02-2399-8518 Abstract: The present study is focused on identifying the most suitable sequence of machine learningbased models and the most promising set of input variables aiming at the estimation of heat transfer in evaporating R134a flows in microfin tubes. Utilizing the available experimental data, dimensionless features representing the evaporation phenomena are first generated and are provided to a machine learning-based model. Feature selection and algorithm optimization procedures are then performed. It is shown that the implemented feature selection method determines only six dimensionless parameters (Sul: liquid Suratman number, Bo: boiling number, Frg: gas Froude number, Rel: liquid Reynolds number, Bd: Bond number, and e/D: fin height to tube’s inner diameter ratio) as the most effective input features, which reduces the model’s complexity and facilitates the interpretation of governing physical phenomena. Furthermore, the proposed optimized sequence of machine learning algorithms (providing a mean absolute relative difference (MARD) of 8.84% on the test set) outperforms the most accurate available empirical model (with an MARD of 19.7% on the test set) by a large margin, demonstrating the efficacy of the proposed methodology. Keywords: machine learning; heat transfer estimation; evaporating flows; R134a; feature selection; relative feature importance 1. Introduction Thermal management of components is a challenging task in many industries such as high-precision manufacturing; miniaturized heating, ventilation, and air conditioning (HVAC) systems; heat pumps; and heat dissipation in electric vehicles. The HVAC sector has long transitioned from using smooth tubes to microfin tubes in condensers and evaporators. In horizontal ducts, the swirled fins help the fluid reach the tube’s upper part, and the dry-out phenomenon occurs at qualities greater than 0.9, which widely compensates for the drawbacks of a larger pressure drop per unit length and higher costs. Evaporation inside a horizontal smooth tube is characterized by a sequence of different flow regimes which changes as the quality shifts from x = 0 to x = 1 and the mass flux changes and is affected by the liquid’s and the vapor’s therma

FAQs

What is the document’s main goal for heat transfer prediction?

To identify effective machine-learning model sequences and input variables for estimating heat transfer during evaporating R134a flow in microfin tubes.

How are input variables constructed and selected?

Dimensionless features representing evaporation phenomena are generated from experimental data, then feature selection reduces them to six most effective dimensionless parameters.

How does the proposed method perform compared with an empirical correlation?

The optimized machine-learning pipeline reaches MARD of 8.84% on the test set, outperforming the best empirical model with MARD of 19.7%.

Aug 4, 2026

Upload by Aria